markb

Well as you can see in the photos, the FOV on the 1/4" camera is noticeably wider than on the 1/3" camera. They were mounted on the same mount for these tests. Both supposedly have 3.6mm lenses. If I'm remembering correctly from my photo hobby days, the normal lens for a 35mm camera is 50mm. But a 50mm lens on a 6x7cm camera is a wide angle. If CCD size works the same way, the 3.6mm should give me a wider field of view with the larger "film" size. Since it doesn't, either the specs are wrong or something else is involved. (I can see the "3.6" on the smaller camera's lens, but the lens on the larger camera is unmarked, so maybe it's really longer.) Mark

Rory, Thanks for your comments. If resolution is the pixel size (512x492), yes that's the same. I thought that 520 vs. 420 TVL meant higher resolution. And shouldn't the larger CCD also give a better image? No doubt. If you or anyone has a better recommendation in the $40 - $90 range, I'd be glad to hear it. I'm still in my 30-day return policy on the 810Bs. Finally read up on what an I/R cut filter is, and I think I see what you mean. But the lack of an I/R filter just means poor color, right? I can live with a washed-out image, but I want it to be as sharp as possible, day or night. Thanks for those instructions. There's a sticker so they'll know if they have been opened. If I decide to keep the cams, I'll probably try focusing them. Good thought. I checked the specs for this Intel PC430 webcam. It says that the "Image Sensor" is "Progressive scan CCD, 640 x 480." But then it says that the "Resolutions (native)" are "40 x 480, 320 x 240, 160 x 120." So maybe the 640x480 is interlaced. Thanks again for the time you take to help out us newbies! Mark

Here is a daylight shot from Camera #2, Mitar MT-306: Same basic scene from Camera #3, Mitar MT-810B: Now back to Camera #2, Mitar MT-306, with digital zoom: And Camera #3, Mitar MT-810B, with digital zoom: These still look fuzzy and washed out to me compared to Rory's jeep picture posted a few days ago, also from a "budget color camera." Maybe my USB converter is the weak link here? Does anyone have a USB converter that they recommend? Thanks, Mark

These samples were captured from live (not recorded) feed the DVR's video out port using AmCap and my old USB webcam converter. IR Image from Camera #2, Mitar MT-306: IR Image from Camera #3, Mitar MT-810B: Is it just me, or does the "better" MT-810B cam look fuzzier? The holder for the towel bar has a crack in it--visible in the first pic but not the second. Does the MT-810B need focusing? The instructions say not to open the camera. Mark

Hi, Thought I'd start a new thread to share some sample images from two Mitar cameras. Continuing the numbering from the previous thread, Camera #2 is a Mitar MT-306, OEM'd as GS-306. Camera #3 is a newer Mitar MT-810B (according to the box--hard to find online), OEM'd as GS-306B. The main difference is in sensor size and lines of resolution. Here are the specs: ------------------------------------ Camera #2 - Mitar MT-306 ------------------------------------ Day & Night Outdoor CCD Camera (Waterproof) • Built-in infrared LED, may work under zero degree of illumination environment • The infrared project distance is between 12~18 meters (40 ~ 60 ft) • Using waterproof metal crust (IP67 standard) • Automatic backlight compensatory function • Automatic tracking white balance Model Name GS306 Pick-Up Device 1/4" Sony CCD Number of Pixels 512(H) x 492(V) System of Signal NTSC/PAL Horizontal Resolution 420TV Lines Minimum Illumination 0 Lux Lens CS Mount, 3.6mm (90°) Backlight Compensation On/Off Electronic Shutter 1/50(1/60)~1/100,000sec White Balance Auto S/N Ratio >48db Gamma Correction >0.45 Operation Temperature -20°C~50°C (-4°F ~ 122°F) Sync System Internal Video Output ≤1.2Vp-p/75Ω Power Source Supply DC12V, ≥750mA Weight Approx. 2 lb Dimension 102x58 mm (4x2.3 in) Send private message ------------------------------------ Camera #3 - Mitar MT-810B ------------------------------------ Day & Night Outdoor CCD Camera (Waterproof) • High Resolution Color IR LED • Built-in infrared LED, may work under zero degree of illumination environment • The infrared project distance is between 25 ~ 30 meters (75 ~ 90 ft) • Using waterproof metal crust (IP67 standard) • Automatic backlight compensatory function • Automatic tracking white balance Model Name GS306B Pick-Up Device 1/3" Sony CCD Number of Pixels 512(H) x 492(V) System of Signal NTSC/PAL Horizontal Resolution 520TV Lines Minimum Illumination 0 Lux Lens CS Mount, 3.6mm (90°) Backlight Compensation On/Off Electronic Shutter 1/50(1/60)~1/100,000sec White Balance Auto S/N Ratio >48db Gamma Correction >0.45 Operation Temperature -20°C~50°C (-4°F ~ 122°F) Sync System Internal Video Output ≤1.2Vp-p/75Ω Power Source Supply DC12V, ≥750mA Weight Approx. 2 lb Dimension 102x58 mm (4x2.3 in) Samples in next post. Mark

I've been having the same issue for the last several days. When I try to upload an attachment (not just a link to an off-site image), I keep getting the message "Could not access ftp directory: 'attachments/thumbs'. Please check your FTP Settings." Anyone else having this trouble? Is the forum out of space? Thanks, Mark

It's an AV Tech 760, the one with a network port but no USB or CD/DVD. The manual covers five models and refers to this as "Model 3". In the Product Description, under Video Compression Format, it says: "Frame: MJPEG; CIF: MPEG4". The CIF quality seems much worse than Frame when paused. Maybe CIF does a better job of capturing motion; otherwise, I'm not sure why it's in there at all. Mark

No worries; I appreciate your help. I uninstalled the Intel Create and Share, and while that leaves behind a basic driver, it doesn't seem to allow selecting the video in port on the webcam. So AmCap won't work unless I re-install all the Intel stuff. Setting aside capture, just eyeballing the video playback on my cheapo b/w monitor, I can see just as much distortion in the IR images as I saw in the captured images that I already posted. Conversely, watching the _live_ monitor (even hooked up through the DVR), the IR image looks sharper than during playback, and the "color" image looks very sharp. So I think the monitor and camera are doing pretty well, and the captures are representing them accurately. It's the DVR that is sucking out the most quality. What settings do you use for AV Tech recording? I have mine set to Frame mode, Best quality, 30 fps when detecting motion. I may add 7 fps for "manual" record (which basically runs all the time), but the problem doesn't seem to be the frame rate, but rather the quality of each frame. All of this is with one camera--will it bog down if I connect more? I wonder if increasing the contrast setting in the DVR for this camera would help. Unfortunately, there's no way to do that for IR mode only. The other camera should arrive tomorrow (today). More to play with. Mark

AmCap compression was set to None. I don't see a Quality setting.

That AmCap is a svelt little program. It'd be great if I could use that instead of the Intel bloatware. I'll have to see if it can access the video in port of the Intel webcam after I uninstall the Intell Create and Share software. Tried to upload the same image captured through AmCap as a still capture, but I'm getting an error after I click on "Add Attachment": "Could not access ftp directory: 'attachments/thumbs'. Please check your FTP Settings." Anyway, the version from the Intel software looks just a little bit better to me. It's also 70KB instead of AmCap's 25KB. Mark

Well I have an old Intel USB webcam here that allows hooking up an external video source and capturing images. Not an ideal solution, but I thought I'd see how it compares to what I posted the other day. First to clarify, the snapshots posted above were captured from the DVR software, but they were captured during paused DVR playback, not after saving the file on the computer. The white time stamp is the time of the capture; the yellow timestamp is the time the image was recorded. I connected the camera directly to the USB interface and captured a couple new shots, #1 and #2 below. Then I went back to that same frame on the DVR and captured it directly from the DVR to the PC via the USB interface. That's #3 below. So it looks to me like the camera by itself itself is decent with ambient lighting, and almost okay with IR. The DVR recording loses quality (though it actually preserves motion pretty well). Playing back the recording through the web interface makes an IR picture almost useless. I can live with the understanding that web playback is poor as long as I can capture acceptable quality, should it ever be needed, directly from the DVR. At this point it's borderline whether this DVR is good enough even for that. (By the way, anyone ever try a DigiMerge DGR204 DVR? I could probably still get one of those instead--see my post on the DVR forum.) I'll see what I can learn (and afford) about true day/night cameras. I think there are some posts on b/w cameras too. I'd rather have good nighttime quality than daytime color. Thanks for your help, Mark

Rory, So am I getting the message that you prefer PC-based DVR for a low-end solution? Here is a pic of the camera cables. The gray is a power/video cable that I was worried is too thin. The black is the old Thin Ethernet network cable, labeled RG58A/U, which for the video alone is as thick as the gray cable with both power and video. A penny for size comparison... Mark

My bad, the MT-306 is camera #2 above, with the 1/4" CCD. I've got one of those on order. The pictures are from a larger (1/3" CCD) MT232A. You can find that on the procamcctv site as well (I can't post URLs yet). The description on their site seems to imply a varifocal lens but mine is fixed at 8mm, which I've found to be too long for most locations around the house. (If I were to mount it on the garage, I could only monitor the far end of my driveway.) I guessed at the "Mitar" name from the printing on the side of the cameras at those web sites. Thanks everyone for the suggestions. I'll play around and see how far I get. The purpose here is not live observation but recording stuff while I'm away, so it has to work through the DVR. I do have an inexpensive CCTV monitor (Lorex B/W). A paused frame looks a wee bit better on that than through the software. The software is pretty good in terms of functionality (it's Java by the way, works fine under XP and Vista, and the remote viewer works in Firefox). Probably the video streaming server, which is in the DVR, isn't the best. I'm also testing cables. Is a BNC cable supposed to be shielded? The 75-foot video/power cable I bought from the supplier is very thin. The IR image through that cable was extremely blocky (I think that's called "pixelated"?). I dug out an old Thin Ethernet computer networking cable, maybe 25 feet long, and used that for the sample pictures. Something like a Belkin F3K101-25-E. More when I can get time to test some of your suggestions... Mark

Attached are three captures from Camera #1 above (I think it's a Mitar MT-306) as recorded on my new AV Tech 760 and played back through the included software. All three are taken indoors with at a distance of about 10 feet. One is taken with daylight illumination, one with incandescent from an overhead light, and one with the camera's built-in IR. Is this considered acceptable quality? I'm concerned that especially with the IR, facial features are almost indiscernable. Is there anything I can do to enhance this? I'm already recording at "Best" quality and the maximum 30 fps in Frame mode (MJPEG). There's also a CIF mode up to 120 ips, but that seems to only record the 4-way split of the DVR, and the playback quality is not as good. This camera does in fact throw good IR all the way across my back yard (~80+ feet). But if you can't identify people at 10 feet, what's the point? Mark San Diego, CA

I'm looking at getting an AVTECH AVC760 or a Digimerge DGR204. Specs look similar, except that the Digimerge holds two HDs. Since the Digimerge is discontinued, it is available at a similar price. Normally I would lean towards the product that is currently in production, but maybe the Digimerge is a more solid product and would outweigh that consideration? Any thoughts? Thanks, Mark